Photon Conversion Classi?cation by Boosting Decision Trees

Abstract

The ALICE detector located at CERN studies subatomic particles produced in heavy-ion collisions. These collisions generate enormous amounts of particles including photons. The data gathered from these collisions is contaminated with background. This research focuses on generating a viable and efficient machine-learning algorithm for selecting photon conversions and discriminating them from background. The method used is that of the boosted decision tree (BDT). A Monte Carlo simulation is used to train and test the BDT and afterwards to test the performance of the BDT. The Monte Carlo simulation consists of data taken from a simulated collision of 40%-60% centrality and a center of mass energy of 2.76$ TeV (Tera electron Volts).